can-i-run-this-llm

Can I run this LLM?

Can a NVIDIA RTX 5090 (32GB) run Devstral Small 2 24B?

Mistral · See model card · estimated for a 32.0 GB machine at 4K context.

Yes — it runs great.

Best fit: Devstral Small 2 24B at Q6_K · 19.4 GB · 56–84 tok/s on a NVIDIA RTX 5090 (32GB).

Every quantization on this machine

QuantSizeSpeedFit
Q2_K8.9 GB100–151 tok/sRuns great
Q3_K_M11.5 GB84–126 tok/sRuns great
Q4_K_M14.3 GB71–107 tok/sRuns great
Q5_K_M16.8 GB63–94 tok/sRuns great
Q6_K19.4 GB56–84 tok/sRuns great
Q8_025.1 GB45–68 tok/sTight / slow
FP1647.2 GB26–39 tok/sWon't fit

Ran Devstral Small 2 24B on a NVIDIA RTX 5090 (32GB)? Add your real numbers →